© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7510
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on two‐character tokens (often uppercase or mixed‐case) that correspond to abbreviations, initials, or short acronyms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    😂
    0.74
     keter
    0.66
    นี่
    0.65
    ណ៍
    0.64
     sitä
    0.63
    jší
    0.63
    hui
    0.63
     ilmiah
    0.61
    OWER
    0.61
    ╯
    0.61
    POSITIVE LOGITS
    ̧
    0.86
    ourt
    0.81
    ouncil
    0.80
    oyote
    0.77
    ourses
    0.76
    erca
    0.75
    ritical
    0.75
    ounters
    0.74
    ourse
    0.74
    ruitment
    0.73
    Activations Density 0.959%

    No Known Activations